The objective of this proposal is to mount a major research program on adjustment to chronic disease. The research will focus on how social support and feelings of control facilitate psychological and health outcomes among chronically ill patients. Three specific research projects are proposed. The first evaluates the effects of a control-restoring intervention (CPR training) on high-risk cardiac patients and their families. Previously, we found that CPR training of family members raised anxiety and depression among patients. The present study, a clinical trial funded by NHLBI, compares a no-treatment control group with two conditions designed to offset these deleterious effects: CPR training with risk factor/warning signs training and CPR training with a social support intervention. The study also identifies family dynamics introduced by the intervention. The results have policy implications concerning whether CPR training for families of high-risk patients reduces mortality, improves responses to potential sudden death events, and does so without undue psychological costs. The second project, funded by NCI, explores social support needs and the use of social support groups among cancer patients (N=650) using interview and questionnaire data. Previous research indicates that cancer patients experience difficulties in social support, yet few turn to social support groups. Male, low-SES, and minority patients are particularly unlikely to do so. This project assesses the type and frequency of psychosocial problems among patients as well as reasons for not joining, dropping out of, or faithfully attending support groups. Part two of this project evaluates a support group format designed to appeal especially to male, working class, and minority cancer patients against a more traditional support group format in an experimental, longitudinal design. The project will also generate guidelines for social support groups that appeal maximally to target populations. The third project explores the implications of failed efforts to exert control, specifically, what happens when patients assume responsibility for medical decisions that ultimately fail? Previous research from cognitive dissonance, reactance and learned helplessness generate contradictory predictions. Using quasi-experimental longitudinal data from patients whose treatments either succeed or fail (kidney transplantation patients, chemotherapy patients), we will build an integrative model of reactions to loss of control. The results also have important implications for the informed consent procedure.